Fault Detection in Three-Phase Power Systems Using Artificial Neural Networks (ANN) / Deep Learning in MATLAB

In this Phase I study, the conceptual framework and design approach for fault detection in three-phase power systems using Artificial Neural Networks (ANN) and Deep Learning techniques have been successfully proposed and documented. The project establishes a strong theoretical foundation for developing an intelligent fault detection system capable of identifying and classifying various electrical faults based on RYB phase voltage and current data. The report presented an in-depth analysis of: The data acquisition and pre-processing methodologies for RYB three-phase systems. Simulation-based generation of fault scenarios in MATLAB/Simulink. The design architecture of ANN and advanced deep learning models suitable for time-series fault classification. A step-by-step MATLAB implementation plan including data simulation, feature extraction, model training, and validation. Through this study, the feasibility and advantages of applying AI-based models in power system protection have been clearly established. The proposed design offers high accuracy, adaptability, and automation in detecting faults, thereby improving the reliability and efficiency of the electrical system.   However, as this phase focuses primarily on system design, model selection, and theoretical validation, actual model training, testing, and real-time implementation are planned for the second phase of the project. This will include dataset generation through simulation, ANN model development, performance evaluation, and real-time integration. Thus, the current work serves as a blueprint and preparatory stage for the next phase, ensuring that all design components are conceptually validated and aligned with the project objectives.  

[1]       Abhishek Jadhavar, Sandhyarani Gavali, Aashlesha Chavan, and Sayyad A.P., “Transmission Line Fault Detection System,” V.D.F. School of Engineering and Technology, Latur, India. [2]     Yıldırım Özüpak, “Machine Learning-Based Fault Detection in Transmission Lines: A Comparative Study with Random Search Optimization,” Dicle University, Türkiye. [3]      Nadeem Ahmed Tunio, Ashfaque Ahmed Hashmani, Suhail Khokhar, Mohsin Ali Tunio, and Muhammad Faheem, “Fault Detection and Classification in Overhead Transmission Lines through Comprehensive Feature Extraction Using Temporal Convolution Neural Network,” Mehran University of Engineering & Technology, Pakistan, and University of Vaasa, Finland. [4]     Gaurav Jichkar, Rahul Mohurle, and Vikas Kalambhe, “Electrical Fault Detection in Overhead Transmission Line,” Guru Nanak Institute of Engineering and Technology, Nagpur, India. [5]   Abhinay Vijayrao Sahare, Shreyas Sunil Patil, Sejal Arun Shete, Rutuja Sanjay Kandekar, and Minaj S. Shikalgar, “Transmission Line Fault Monitoring System Using IoT – A Project Review Paper,” Kolhapur Institute of Technology’s Autonomous College of Engineering, Kolhapur, India. [6]      R.A. Mahmoud, “Transmission Line Fault Detection and Classification Using New Tripping Characteristics Based on Statistical Coherence for Current Measurements.”